div <- read.csv(file="/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/GenomeScanInput.inclInvariant.MAC4.csv")
div.outliers.USUK <- read.csv(file="/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/FstOutliers.USUK.csv")
div.outliers.AUUK <- read.csv(file="/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/FstOutliers.AUUK.csv")
div.hifst.AUUK <- div[which(div$WEIGHTED_FST_AUUK > 0.1),]
div.hifst.UKUS <- div[which(div$WEIGHTED_FST_UKUS > 0.1),]

Histograms of FST, pi

What’s the statistical distribution of these values?

descdist(div$WEIGHTED_FST_AUUK)

## summary statistics
## ------
## min:  0   max:  0.38607 
## median:  0.0246106 
## mean:  0.02909312 
## estimated sd:  0.02610897 
## estimated skewness:  1.999146 
## estimated kurtosis:  11.03661
descdist(div$WEIGHTED_FST_UKUS)

## summary statistics
## ------
## min:  0   max:  0.291877 
## median:  0.0113938 
## mean:  0.01655209 
## estimated sd:  0.02024904 
## estimated skewness:  2.727796 
## estimated kurtosis:  18.24573
descdist(div$WEIGHTED_FST_USAU)

## summary statistics
## ------
## min:  0   max:  0.390333 
## median:  0.0310777 
## mean:  0.03892551 
## estimated sd:  0.03806378 
## estimated skewness:  3.165728 
## estimated kurtosis:  19.14777
lab.AU <- rep("AU.UK",length(div$WEIGHTED_FST_AUUK))
lab.US <- rep("UK.US",length(div$WEIGHTED_FST_UKUS))
Fst.group <- c(lab.AU,lab.US)
Fst.hist.data <- c(div$WEIGHTED_FST_AUUK,div$WEIGHTED_FST_USUK)
Fst.hist <- data.frame(Fst = Fst.hist.data, population = Fst.group)
pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/HistDensity_Fst.pdf",width=4,height=3)
ggplot(Fst.hist, aes(x=Fst, y=..density.., fill=population)) +
  theme_classic() +
  geom_density(alpha=0.5,lwd=0.5) +
  scale_fill_manual(values=c("#F2C14E","#2c81a8")) + xlim(0,0.5) + 
  xlab("Fst") + labs(fill="Population") + 
  geom_vline(xintercept=0.03,colour=alpha("#F2C14E"),linetype="dashed", size=1) +
  geom_vline(xintercept=0.01,colour=alpha("#2c81a8"),linetype="dashed", size=1) +
  geom_vline(xintercept=0.08,colour=alpha("gray50"),linetype="dotted", size=0.5)
dev.off()
## quartz_off_screen 
##                 2
ggplot(Fst.hist, aes(x=Fst, y=..density.., fill=population)) +
  theme_classic() +
  geom_density(alpha=0.5,lwd=0.5) +
  scale_fill_manual(values=c("#F2C14E","#2c81a8")) + xlim(0,0.5) + 
  xlab("Fst") + labs(fill="Population") + 
  geom_vline(xintercept=0.03,colour=alpha("#F2C14E"),linetype="dashed", size=1) +
  geom_vline(xintercept=0.01,colour=alpha("#2c81a8"),linetype="dashed", size=1) +
  geom_vline(xintercept=0.08,colour=alpha("gray50"),linetype="dotted", size=0.5)

ggplot(data=div) +
  geom_point(aes(x=div$WEIGHTED_FST_UKUS, y=div$PI_US),col="#2c81a8",cex=0.7) +
  #xlab("Fst (Native vs. Invasive)") + ylab("Pi Invasive") +
  xlab("") + ylab("") +
  stat_smooth(aes(x=WEIGHTED_FST_UKUS, y=PI_US),method="loess",col="black",lwd=0.5) +
  xlim(0,0.31) + ylim(0,0.04) + theme_classic()
## Warning: Use of `div$WEIGHTED_FST_UKUS` is discouraged. Use `WEIGHTED_FST_UKUS`
## instead.
## Warning: Use of `div$PI_US` is discouraged. Use `PI_US` instead.
## `geom_smooth()` using formula 'y ~ x'
## Warning: Computation failed in `stat_smooth()`:
## 'Calloc' could not allocate memory (18446744072404226048 of 8 bytes)

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/FstPi.density.USUK.pdf",height=2,width=2)
ggplot(data=div) +
  geom_point(aes(x=div$WEIGHTED_FST_UKUS, y=div$PI_US),col="#2c81a8",cex=0.7) +
  #xlab("Fst (Native vs. Invasive)") + ylab("Pi Invasive") +
  xlab("") + ylab("") +
  stat_smooth(aes(x=WEIGHTED_FST_UKUS, y=PI_US),method="loess",col="black",lwd=0.5) +
  xlim(0,0.31) + ylim(0,0.04) + theme_classic()
## Warning: Use of `div$WEIGHTED_FST_UKUS` is discouraged. Use `WEIGHTED_FST_UKUS`
## instead.

## Warning: Use of `div$PI_US` is discouraged. Use `PI_US` instead.
## `geom_smooth()` using formula 'y ~ x'
## Warning: Computation failed in `stat_smooth()`:
## 'Calloc' could not allocate memory (18446744072404226048 of 8 bytes)
dev.off()
## quartz_off_screen 
##                 2
ggplot(data=div) +
  geom_point(aes(x=div$WEIGHTED_FST_AUUK, y=div$PI_AU),col="#F2C14E",cex=0.7) +
  xlim(0,0.31) + ylim(0,0.04) + theme_classic() +
  #xlab("Fst (Native vs. Invasive)") + ylab("Pi Invasive") +
  stat_smooth(aes(x=WEIGHTED_FST_AUUK, y=PI_AU),method="loess",col="black",lwd=0.5) +
  xlab("") + ylab("")
## Warning: Use of `div$WEIGHTED_FST_AUUK` is discouraged. Use `WEIGHTED_FST_AUUK`
## instead.
## Warning: Use of `div$PI_AU` is discouraged. Use `PI_AU` instead.
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 3 rows containing non-finite values (stat_smooth).
## Warning: Computation failed in `stat_smooth()`:
## 'Calloc' could not allocate memory (18446744072403847168 of 8 bytes)
## Warning: Removed 3 rows containing missing values (geom_point).

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/FstPi.density.AUUK.pdf",height=2,width=2)
ggplot(data=div) +
  geom_point(aes(x=div$WEIGHTED_FST_AUUK, y=div$PI_AU),col="#F2C14E",cex=0.7) +
  xlim(0,0.31) + ylim(0,0.04) + theme_classic() +
  #xlab("Fst (Native vs. Invasive)") + ylab("Pi Invasive") +
  stat_smooth(aes(x=WEIGHTED_FST_AUUK, y=PI_AU),method="loess",col="black",lwd=0.5) +
  xlab("") + ylab("")
## Warning: Use of `div$WEIGHTED_FST_AUUK` is discouraged. Use `WEIGHTED_FST_AUUK`
## instead.
## Warning: Use of `div$PI_AU` is discouraged. Use `PI_AU` instead.
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 3 rows containing non-finite values (stat_smooth).
## Warning: Computation failed in `stat_smooth()`:
## 'Calloc' could not allocate memory (18446744072403847168 of 8 bytes)
## Warning: Removed 3 rows containing missing values (geom_point).
dev.off()
## quartz_off_screen 
##                 2
summary(div$PI_AU)
##      Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 
## 1.933e-05 3.302e-03 5.088e-03 5.060e-03 6.732e-03 3.313e-02
summary(div$PI_UK)
##      Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 
## 9.450e-06 3.507e-03 5.494e-03 5.461e-03 7.260e-03 3.319e-02
summary(div$PI_US)
##      Min.   1st Qu.    Median      Mean   3rd Qu.      Max. 
## 3.758e-05 3.487e-03 5.438e-03 5.412e-03 7.185e-03 3.379e-02
descdist(div$PI_US)

## summary statistics
## ------
## min:  3.75824e-05   max:  0.0337912 
## median:  0.00543846 
## mean:  0.005412495 
## estimated sd:  0.002682266 
## estimated skewness:  0.2262385 
## estimated kurtosis:  3.378289
#norm.piUS.fit <- fitdist(div$PI_US,distr="norm")
#summary(norm.piUS.fit)
descdist(div$PI_UK)

## summary statistics
## ------
## min:  9.4515e-06   max:  0.0331918 
## median:  0.00549351 
## mean:  0.005461298 
## estimated sd:  0.002705316 
## estimated skewness:  0.2234852 
## estimated kurtosis:  3.337609
#norm.piUK.fit <- fitdist(div$PI_UK,distr="norm")
#summary(norm.piUK.fit)
descdist(div$PI_AU)

## summary statistics
## ------
## min:  1.93333e-05   max:  0.0331342 
## median:  0.00508791 
## mean:  0.005060351 
## estimated sd:  0.002513141 
## estimated skewness:  0.2032717 
## estimated kurtosis:  3.405112
#norm.piAU.fit <- fitdist(div$PI_AU,distr="norm")
#summary(norm.piAU.fit)
lab.AU <- rep("AU",length(div$PI_AU))
lab.US <- rep("US",length(div$PI_US))
lab.UK <- rep("UK",length(div$PI_UK))
group <- c(lab.AU,lab.US,lab.UK)
pi.hist.data <- c(div$PI_UK,div$PI_US,div$PI_AU)
pi.hist.lab <- data.frame(pi = pi.hist.data, population = group)
str(pi.hist.lab)
## 'data.frame':    133923 obs. of  2 variables:
##  $ pi        : num  0.00263 0.00299 0.00303 0.00156 0.0015 ...
##  $ population: Factor w/ 3 levels "AU","UK","US": 1 1 1 1 1 1 1 1 1 1 ...
ggplot(pi.hist.lab, aes(x=pi, y=..density.., fill=population)) +
  geom_density(alpha=0.8,lwd=0.5) + theme_classic() +
  scale_fill_manual(values=c("black","#2c81a8","#F2C14E")) + xlim(-0.0001,0.02) + 
  xlab("Pi") + labs(fill="Population") +
  geom_vline(xintercept=mean(div$PI_AU),colour=alpha("#F2C14E"),linetype="dashed", size=1) +
  geom_vline(xintercept=mean(div$PI_US),colour=alpha("#2c81a8"),linetype="dashed", size=1) +
  theme(legend.position="none")
## Warning: Removed 5 rows containing non-finite values (stat_density).

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/HistDensity_Pi.pdf",width=4,height=3)
ggplot(pi.hist.lab, aes(x=pi, y=..density.., fill=population)) +
  geom_density(alpha=0.8,lwd=0.5) + theme_classic() +
  scale_fill_manual(values=c("black","#2c81a8","#F2C14E")) + xlim(-0.0001,0.02) + 
  xlab("Pi") + labs(fill="Population") +
  geom_vline(xintercept=mean(div$PI_US),colour=alpha("#2c81a8"),linetype="dashed", size=1) +
  geom_vline(xintercept=mean(div$PI_AU),colour=alpha("#F2C14E"),linetype="dashed", size=1) +
  theme(legend.position="none")
## Warning: Removed 5 rows containing non-finite values (stat_density).
dev.off()
## quartz_off_screen 
##                 2

Average nucleotide diversity for both invasions is the same (0.003). There are two vertical lines overlaid in the plot above.

ggplot(data=div) +
  geom_point(aes(x=PI_UK, y=PI_US),col="#2c81a8",cex=0.7) +
  xlab("") + ylab("") + xlim(0,0.02) + ylim(0,0.02) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_US),span=0.2,method="loess",col="black",lwd=0.5)
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).
## Warning: Removed 2 rows containing missing values (geom_point).

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/Pi_USvsUK.pdf",width=2,height=2)
ggplot(data=div) +
  geom_point(aes(x=PI_UK, y=PI_US),col="#2c81a8",cex=0.7) +
  xlab("") + ylab("") + xlim(0,0.02) + ylim(0,0.02) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_US),span=0.2,method="loess",col="black",lwd=0.5)
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).

## Warning: Removed 2 rows containing missing values (geom_point).
dev.off()
## quartz_off_screen 
##                 2
ggplot(data=div.outliers.USUK) +
  geom_point(aes(x=PI_UK, y=PI_US),col="#2c81a8",cex=0.7) +
  xlab("") + ylab("") + xlim(0,0.01) + ylim(0,0.01) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_US),span=0.2,method="loess",col="black",lwd=0.5)
## `geom_smooth()` using formula 'y ~ x'

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/PiOutliers_USvsUK.pdf",width=2,height=2)
ggplot(data=div.outliers.USUK) +
  geom_point(aes(x=PI_UK, y=PI_US),col="#2c81a8",cex=0.7) +
  xlab("") + ylab("") + xlim(0,0.01) + ylim(0,0.01) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_US),span=0.2,method="loess",col="black",lwd=0.5)
## `geom_smooth()` using formula 'y ~ x'
dev.off()
## quartz_off_screen 
##                 2
ggplot(data=div) +
  geom_point(aes(x=PI_UK, y=PI_AU),col="#F2C14E",cex=0.7) +
  xlab("") + ylab("") +
  xlim(0,0.02) + ylim(0,0.02) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_AU),span=0.2,method="loess",col="black",lwd=0.5) 
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).

## Warning: Removed 2 rows containing missing values (geom_point).

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/Pi_AUvsUK.pdf",width=2,height=2)
ggplot(data=div) +
  geom_point(aes(x=PI_UK, y=PI_AU),col="#F2C14E",cex=0.7) +
  xlab("") + ylab("") +
  xlim(0,0.02) + ylim(0,0.02) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_AU),span=0.2,method="loess",col="black",lwd=0.5) 
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 2 rows containing non-finite values (stat_smooth).

## Warning: Removed 2 rows containing missing values (geom_point).
dev.off()
## quartz_off_screen 
##                 2
ggplot(data=div.outliers.AUUK) +
  geom_point(aes(x=PI_UK, y=PI_AU),col="#F2C14E",cex=0.7) +
  xlab("") + ylab("") +
  xlim(0,0.01) + ylim(0,0.01) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_AU),span=0.2,method="loess",col="black",lwd=0.5) 
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 1 rows containing missing values (geom_smooth).

dev.off()
## null device 
##           1
pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/Pi_AUvsUK.pdf",width=2,height=2)
ggplot(data=div.outliers.AUUK) +
  geom_point(aes(x=PI_UK, y=PI_AU),col="#F2C14E",cex=0.7) +
  xlab("") + ylab("") +
  xlim(0,0.01) + ylim(0,0.01) + theme_classic() +
  theme(axis.text=element_text(size=7,colour="black")) +
  stat_smooth(aes(x=PI_UK, y=PI_AU),span=0.2,method="loess",col="black",lwd=0.5) 
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 1 rows containing missing values (geom_smooth).
dev.off()
## null device 
##           1
descdist(div$TajimaD_US)

## summary statistics
## ------
## min:  -2.1488   max:  3.16509 
## median:  0.7354 
## mean:  0.7200025 
## estimated sd:  0.2762649 
## estimated skewness:  -1.283181 
## estimated kurtosis:  15.94083
descdist(div$TajimaD_UK)

## summary statistics
## ------
## min:  -2.2276   max:  3.23445 
## median:  0.726854 
## mean:  0.7107561 
## estimated sd:  0.2594173 
## estimated skewness:  -0.4769201 
## estimated kurtosis:  14.32178
descdist(div$TajimaD_AU)

## summary statistics
## ------
## min:  -2.33194   max:  3.16824 
## median:  0.792182 
## mean:  0.7784472 
## estimated sd:  0.2745281 
## estimated skewness:  -0.6274247 
## estimated kurtosis:  14.38053
descdist(div.outliers.USUK$TajimaD_US)

## summary statistics
## ------
## min:  -2.1488   max:  2.36598 
## median:  0.798903 
## mean:  0.6287185 
## estimated sd:  0.7633274 
## estimated skewness:  -1.270981 
## estimated kurtosis:  4.923513
descdist(div.outliers.AUUK$TajimaD_AU)

## summary statistics
## ------
## min:  -2.30499   max:  2.47406 
## median:  0.502038 
## mean:  0.6191664 
## estimated sd:  0.5889809 
## estimated skewness:  -1.134327 
## estimated kurtosis:  8.187706

Differences in diversity

lab.AU <- rep("AU",length(div$piUK.piAU))
lab.US <- rep("US",length(div$piUK.piUS))
group <- c(lab.AU,lab.US)
pi.hist.data <- c(div$piUK.piUS,div$piUK.piAU)
pi.hist.lab <- data.frame(pi = pi.hist.data, population = group)
str(pi.hist.lab)
## 'data.frame':    89282 obs. of  2 variables:
##  $ pi        : num  9.82e-05 -3.13e-04 1.96e-04 2.03e-04 3.89e-04 ...
##  $ population: Factor w/ 2 levels "AU","US": 1 1 1 1 1 1 1 1 1 1 ...
ggplot(pi.hist.lab, aes(x=pi, y=..density.., fill=population)) +
  geom_vline(xintercept=0,colour="black",size=0.5) +
  geom_vline(xintercept=0.0005,colour="black",size=0.5) +
  geom_vline(xintercept=-0.0005,colour="black",size=0.5) +
  geom_density(alpha=0.8,lwd=0.5) + theme_classic() +
  scale_fill_manual(values=c("#2c81a8","#F2C14E")) + 
  xlab("Difference in pi") + labs(fill="Population") +
  geom_vline(xintercept=mean(div$piUK.piAU),colour=alpha("#F2C14E"), size=1) +
  geom_vline(xintercept=mean(div$piUK.piUS),colour=alpha("#2c81a8"), size=1) +
  theme(legend.position="none")

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/HistDensity_PiDifference.pdf",width=4,height=3)
ggplot(pi.hist.lab, aes(x=pi, y=..density.., fill=population)) +
  geom_vline(xintercept=0,colour="black",size=0.5) +
  geom_vline(xintercept=0.0005,colour="black",size=0.5) +
  geom_vline(xintercept=-0.0005,colour="black",size=0.5) +
  geom_density(alpha=0.8,lwd=0.5) + theme_classic() +
  scale_fill_manual(values=c("#2c81a8","#F2C14E")) + 
  xlab("Difference in pi") + labs(fill="Population") +
  geom_vline(xintercept=mean(div$piUK.piAU),colour=alpha("#F2C14E"), size=1) +
  geom_vline(xintercept=mean(div$piUK.piUS),colour=alpha("#2c81a8"), size=1) +
  theme(legend.position="none")
dev.off()
## quartz_off_screen 
##                 2

Plotting high-fst (FST > 0.1) windows only

lab.AU <- rep("AU",length(div.hifst.AUUK$piUK.piAU))
lab.US <- rep("US",length(div.hifst.UKUS$piUK.piUS))
group <- c(lab.AU,lab.US)
pi.hist.data <- c(div.hifst.UKUS$piUK.piUS,div.hifst.AUUK$piUK.piAU)
pi.hist.lab <- data.frame(pi = pi.hist.data, population = group)
str(pi.hist.lab)
## 'data.frame':    1323 obs. of  2 variables:
##  $ pi        : num  5.10e-04 5.84e-04 3.93e-04 -4.92e-04 3.41e-05 ...
##  $ population: Factor w/ 2 levels "AU","US": 1 1 1 1 1 1 1 1 1 1 ...
ggplot(pi.hist.lab, aes(x=pi, y=..density.., fill=population)) +
  geom_vline(xintercept=0,colour="black",size=0.5) +
  geom_vline(xintercept=0.0005,colour="black",size=0.5) +
  geom_vline(xintercept=-0.0005,colour="black",size=0.5) +
  geom_density(alpha=0.8,lwd=0.5) + theme_classic() +
  scale_fill_manual(values=c("#2c81a8","#F2C14E")) + 
  xlab("Difference in pi") + labs(fill="Population") +
  geom_vline(xintercept=mean(div.hifst.AUUK$piUK.piAU),colour=alpha("#F2C14E"), size=1) +
  geom_vline(xintercept=mean(div.hifst.UKUS$piUK.piUS),colour=alpha("#2c81a8"), size=1) +
  theme(legend.position="none")

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/HistDensity_PiDifference_HiFst.pdf",width=4,height=3)
ggplot(pi.hist.lab, aes(x=pi, y=..density.., fill=population)) +
  geom_vline(xintercept=0,colour="black",size=0.5) +
  geom_vline(xintercept=0.0005,colour="black",size=0.5) +
  geom_vline(xintercept=-0.0005,colour="black",size=0.5) +
  geom_density(alpha=0.8,lwd=0.5) + theme_classic() +
  scale_fill_manual(values=c("#2c81a8","#F2C14E")) + 
  xlab("Difference in pi") + labs(fill="Population") +
  geom_vline(xintercept=mean(div.hifst.AUUK$piUK.piAU),colour=alpha("#F2C14E"), size=1) +
  geom_vline(xintercept=mean(div.hifst.UKUS$piUK.piUS),colour=alpha("#2c81a8"), size=1) +
  theme(legend.position="none")
dev.off()
## quartz_off_screen 
##                 2

Are regions with novel pi also highly differentiated? Expect this scatterplot to look bimodal, where shifts in diversity in either direction led to differentiation between populations.

ggplot(data=div) +
  geom_point(aes(x=WEIGHTED_FST_UKUS, y=piUK.piUS),col="#2c81a8",cex=0.7) +
  #xlab("Fst (Native vs. Invasive)") + ylab("Pi Invasive") +
  xlab("") + ylab("") +
  stat_smooth(aes(x=WEIGHTED_FST_UKUS, y=piUK.piUS),method="loess",col="black",lwd=0.5) +
  xlim(0,0.31) + ylim(-0.002,0.002) + theme_classic()
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 8035 rows containing non-finite values (stat_smooth).
## Warning: Removed 8035 rows containing missing values (geom_point).

ggplot(data=div) +
  geom_point(aes(x=WEIGHTED_FST_AUUK, y=piUK.piAU),col="#F2C14E",cex=0.7) +
  #xlab("Fst (Native vs. Invasive)") + ylab("Pi Invasive") +
  xlab("") + ylab("") +
  stat_smooth(aes(x=WEIGHTED_FST_AUUK, y=piUK.piAU),method="loess",col="black",lwd=0.5) +
  xlim(0,0.31) + ylim(-0.002,0.002) + theme_classic()
## `geom_smooth()` using formula 'y ~ x'
## Warning: Removed 8046 rows containing non-finite values (stat_smooth).
## Warning: Removed 8046 rows containing missing values (geom_point).

ggplot(data=div) +
  geom_point(aes(x=WEIGHTED_FST_UKUS, y=piUK.piUS),col="#2c81a8",cex=0.7) +
  geom_point(aes(x=WEIGHTED_FST_AUUK, y=piUK.piAU),col="#F2C14E",cex=0.7) +
  xlab("Fst (Native vs. Invasive)") + ylab("Pi Native - Pi Invasive") +
  xlim(-0.01,0.41) + ylim(-0.003,0.003) + theme_bw() +
  geom_density_2d(aes(x=WEIGHTED_FST_AUUK, y=piUK.piAU), colour="#ffffff") +
  geom_density_2d(aes(x=WEIGHTED_FST_UKUS, y=piUK.piUS), colour="#2c81a8") +
  guides(col = guide_legend(label = TRUE, label.position = "bottom", 
                           direction = "horizontal"))
## Warning: Removed 5937 rows containing non-finite values (stat_density2d).
## Warning: Removed 6128 rows containing non-finite values (stat_density2d).
## Warning: Removed 6128 rows containing missing values (geom_point).
## Warning: Removed 5937 rows containing missing values (geom_point).

pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/FstPi.density.pdf",height=5,width=5)
ggplot(data=div) +
  geom_point(aes(x=WEIGHTED_FST_UKUS, y=piUK.piUS),col="#2c81a8",cex=0.7) +
  geom_point(aes(x=WEIGHTED_FST_AUUK, y=piUK.piAU),col="#F2C14E",cex=0.7) +
  xlab("Fst (Native vs. Invasive)") + ylab("Pi Native - Pi Invasive") +
  xlim(-0.01,0.41) + ylim(-0.003,0.003) + theme_bw() +
  geom_density_2d(aes(x=WEIGHTED_FST_AUUK, y=piUK.piAU), colour="#ffffff") +
  geom_density_2d(aes(x=WEIGHTED_FST_UKUS, y=piUK.piUS), colour="#2c81a8") +
  guides(col = guide_legend(label = TRUE, label.position = "bottom", 
                           direction = "horizontal"))
## Warning: Removed 5937 rows containing non-finite values (stat_density2d).
## Warning: Removed 6128 rows containing non-finite values (stat_density2d).
## Warning: Removed 6128 rows containing missing values (geom_point).
## Warning: Removed 5937 rows containing missing values (geom_point).
dev.off()
## quartz_off_screen 
##                 2
#pdf("/Users/nataliehofmeister/Documents/Ch3-Global-RESEQ/analysis/R/FstPi.density.USUK.pdf",height=2,width=2)
#dev.off()

What’s going on with diversity & TajD at outliers?

library(fitdistrplus)
descdist(div.outliers.AUUK$PI_UK)

## summary statistics
## ------
## min:  7.12114e-05   max:  0.00997923 
## median:  0.00186572 
## mean:  0.002713309 
## estimated sd:  0.002757332 
## estimated skewness:  1.125455 
## estimated kurtosis:  3.633045
descdist(div.outliers.AUUK$PI_AU)

## summary statistics
## ------
## min:  6.86667e-05   max:  0.00875462 
## median:  0.00168611 
## mean:  0.00240641 
## estimated sd:  0.002410777 
## estimated skewness:  1.034319 
## estimated kurtosis:  3.436267
descdist(div.outliers.USUK$PI_UK)

## summary statistics
## ------
## min:  7.12114e-05   max:  0.00743013 
## median:  0.00226024 
## mean:  0.002638915 
## estimated sd:  0.001907804 
## estimated skewness:  0.3509667 
## estimated kurtosis:  2.017906
descdist(div.outliers.USUK$PI_US)

## summary statistics
## ------
## min:  4.41758e-05   max:  0.00875151 
## median:  0.00255646 
## mean:  0.002632511 
## estimated sd:  0.001969958 
## estimated skewness:  0.4448947 
## estimated kurtosis:  2.243299
#beta.outliers.AUUK <- fitdist(div.outliers.AUUK$PI_AU, "beta")
#summary(beta.outliers.AUUK)

#beta.outliers.USUK <- fitdist(div.outliers.USUK$PI_US, "beta")
#summary(beta.outliers.USUK)

descdist(div.outliers.AUUK$TajimaD_AU)

## summary statistics
## ------
## min:  -2.30499   max:  2.47406 
## median:  0.502038 
## mean:  0.6191664 
## estimated sd:  0.5889809 
## estimated skewness:  -1.134327 
## estimated kurtosis:  8.187706
descdist(div.outliers.AUUK$TajimaD_UK)

## summary statistics
## ------
## min:  -1.29367   max:  2.43202 
## median:  0.627764 
## mean:  0.6065902 
## estimated sd:  0.5453867 
## estimated skewness:  0.296591 
## estimated kurtosis:  3.857947
descdist(div.outliers.USUK$TajimaD_US)

## summary statistics
## ------
## min:  -2.1488   max:  2.36598 
## median:  0.798903 
## mean:  0.6287185 
## estimated sd:  0.7633274 
## estimated skewness:  -1.270981 
## estimated kurtosis:  4.923513
descdist(div.outliers.USUK$TajimaD_UK)

## summary statistics
## ------
## min:  -1.92522   max:  2.58948 
## median:  0.780056 
## mean:  0.8487691 
## estimated sd:  0.5719869 
## estimated skewness:  -0.2475749 
## estimated kurtosis:  4.994879